1. General topics
  3. Tools
  5. Bilayer Convergence Question

normal Bilayer Convergence Question

  • andrewmcd
  • andrewmcd's Avatar Topic Author
  • Visitor
9 years 10 months ago #3868 by andrewmcd
Bilayer Convergence Question was created by andrewmcd
I'm trying to complete the bi-layer tutorial and am having a bit of trouble. I run the files in the spontaneous-assembly folder using the commands copied directly from the website. When run with the initial 500 step limit the minimization function fails to converge. When I increased the step limit to 50000 the minimization function converges at step 3898. When I run the "grompp -f md.mdp -c minimised.gro -p dspc.top -maxwarn 10" then "mdrun -v" commands the simulation also converges on step 3898. However, this takes ~2 min, not the 45 described in the tutorial. When I view the confout.gro file with ngmx it displays a completely unordered set of lipids. Any ideas on where to go with this?

Thank you in advance

Please Log in or Create an account to join the conversation.

More
9 years 10 months ago #3869 by Clement
Replied by Clement on topic Bilayer Convergence Question
What do you call "converged"? When the minimization stops on its own? Most of the time, it isn't required for a reasonable energy minimization of your system; when the highest forces of your system reaches ~10^2 or ~10^3, it's already minimized enough.

And if both simulations stop at step 3898, it's because you use the same .tpr... Add a -o flag to your grompp, and make sure you sure you're using the good parameter file (.mdp).

Please Log in or Create an account to join the conversation.

  • andrewmcd
  • andrewmcd's Avatar Topic Author
  • Visitor
9 years 10 months ago #3885 by andrewmcd
Replied by andrewmcd on topic Bilayer Convergence Question
Thanks for the quick feedback. The minimization makes sense to me. I'm still a bit confused about the -o flag. What I've done so far is download the tutorial, and have run the first part of the bilayer setup by copying in the commands directly. I also tried adding the -o flag, but the results didn't change. Is there something I need to modify from the code listed on the website?

Thank you

Please Log in or Create an account to join the conversation.

  1. General topics
  3. Tools
  5. Bilayer Convergence Question
Time to create page: 0.096 seconds